KR930004754B1 - Method and apparatus for effecting group management of elevators - Google Patents

Abstract

No content.

Description

Military management device of elevator and its management method

1 is a block diagram of an overall group management apparatus for an elevator according to a first embodiment of the present invention.

2 is a block diagram of the military management device of FIG.

3 is a flowchart of the military management program.

4 is a flowchart of an empty car detection program.

5 is a flowchart of the car position detection program of FIG.

6 is a flowchart of the car number prediction program of FIG.

7 is a flowchart of the allocation restriction program of FIG.

8 is a segmentation display of detection.

8 and 10 are diagrams showing the relationship between a particular call and a car position in the first embodiment.

11 is a graph showing a modification of the first embodiment.

12 is a flowchart showing an allocation restriction program according to a second embodiment of the present invention.

13 and 14 are diagrams showing the relationship between the specific call and the car position according to the second embodiment of the present invention.

15 to 22 are diagrams showing the relationship between a specific call of a conventional elevator and a car position.

The present invention relates to an efficient group management apparatus for an elevator and a group management method for selecting an appropriate elevator in response to a call from a platform among a plurality of elevators.

In general, military management operation is performed in a building equipped with a plurality of elevators. A typical example of this military management operation is the allocation method. In the allocation method, when the landing call is registered, the allocation evaluation value for each elevator car (hereinafter abbreviated as car) is calculated and the car of the optimal allocation evaluation value is selected to allocate the car to be served for the landing call. In this allocation method, only a specific car responds to the platform call, so that the driving efficiency can be improved and the waiting time can be shortened. In such an assignment-type military management elevator, it is common to provide arrival forecasts and the like for each car and each direction in the platform of each floor. A guest waiting at the platform on each floor knows which car was allocated in response to a call by forecasting the arrival forecast. Therefore, the waiting guest moves to the door of the hatch door where the forecasted specific car is located and waits.

In the above-mentioned allocation method of the landing call, the allocation evaluation value is calculated to determine the car to be allocated in response to the landing call on the assumption that the current situation is advanced as it is. More specifically, in response to the platform call, the estimated time for each car to arrive at the platform on each floor in succession (hereinafter referred to as the arrival time) and the time elapsed since the landing call was registered (hereinafter referred to as elapsed time) Is obtained based on the current position and direction of each car and on the basis of the registered landing call or car call.

In addition, the estimated waiting time for all currently registered platform calls is calculated by taking into account the estimated time of arrival and the elapsed time. The sum of the predicted waiting time or the sum of the squares of the predicted waiting time is set as the above-mentioned allocation evaluation value, and the car whose minimum allocation evaluation value is minimum is assigned to the landing call. However, this conventional method has many problems. For example, the determination of optimality for the platform call is made based on the current situation. As a result, if the user newly registers the platform call after the assignment is completed, the user may wait for a long time. have.

The above problem will be described in detail with reference to the examples of FIGS. 15 to 18. In FIG. 15, the first car A is driving up in response to the car call 6c on the sixth floor, while the second car B is driving in response to the lift allocation 5UB by assigning it to the upcall 5U on the fifth floor. It is present. In this state, as shown in Fig. 16, the falling call 9d is registered. At this time, if the allocation evaluation value is obtained by the conventional allocation method described above, the car A is assigned to minimize the average waiting time for the falling call 9d of the 9th floor, and the car A allocates the falling allocation (9dA) of the 9th floor. When received, both cars will drive upwards.

In addition, if the rising call (1u) on the first floor is registered, for example, after the falling call (9d) on the 9th floor is registered, the rising call (1u) on the first floor becomes the second call to the cars A and B. However, even if the car is assigned to the upcall (1u), it takes a long time to respond to this, and it will inevitably wait for a long time.

On the other hand, the descending call (9d) on the 9th floor is assigned to the car B, and after about 15 seconds as shown in Fig. 17, the car A is called the empty car on the 6th floor, while the car A is the car call on the 10th floor of the service on the 5th floor (10c). Assume that this is registered. In this case, if the ascending call 1u of the first floor is followed, the waiting car A on the sixth floor is immediately moved to the first floor as shown in FIG. Therefore, the user is not forced to wait for a long time.

As mentioned above, in order to prevent the occurrence of long-term waiting time, many cars are not concentrated in one location even if they are allocated temporarily to the long-term waiting time considering the possibility of the near future car position and the possibility of a certain car being vacant. It is necessary to assign a platform call so as not to. As described below, various proposals have been made for how the platform call can be assigned to prevent many cars from concentrating on one location. However, none of the proposed methods achieved a satisfactory effect. The elevator military management device disclosed in Japanese Patent Application No. 55-32625 is designed to prevent concentration of all cars in one location and to improve driving efficiency. When the landing call is registered, the device uses an allocation method for allocating a car that is to be stopped to the landing near the calling floor. However, such a conventional allocation method also has a long waiting time problem.

This conventional approach only pays attention to the presence of cars that are going to be stopped on the next floor, the time required for the arrival of the car on the floor, the distribution of other registration floors, the timing of each response to these registration floors, and the This is because no proper judgment is made about the presence or absence, the floor on which each car is located, the direction in which each car is to travel, and the like.

The military management method of an elevator disclosed in Japanese Patent Application No. 62-56075 is characterized by a method of waiting a car on the floor where the last passenger got off and assigning it from the car. This allocation method is; Temporarily assign each car sequentially for a new landing call to predict the air gap of this temporarily allocated car; Calculates the pH from the estimated waiting position for each temporary allocated car; The staff is configured to select the allocation car for the landing call based on the evaluation value of each car by setting the dispersion degree as the evaluation value of each temporarily assigned car in such a manner that at least the dispersion degree becomes easier to be allocated.

In the allocation method as described above, even after the operation for the platform call is completed, the car is widely distributed while preventing unnecessary empty cars from being dispersed in the standby state. Therefore, the above-mentioned allocation method has the advantage of achieving the above-described energy savings and not damaging the building occupants. However, as is apparent from the purpose disclosed in Japanese Patent Application Laid-Open No. 62-56076, the allocation method covers the case where one platform call is registered when all cars are stopped by empty cars for infrequent driving such as nighttime. Because of the premise, this allocation method is called platform landings in turn. It is not applicable to the allocation of landing platform calls under the driving state in which the cars are driven in response to each call, thus causing a long waiting time.

The first reason for this problem is that since the allocation method aims at balancing the placement of vinca, it is not considered that each car position other than the above-mentioned temporary allocating car changes with time. As can be seen from the premise of this allocation method, it is not necessary to consider the change of position of other cars. The second reason is that the tentatively assigned car determines the allocation to the platform call by focusing only on the position where the last passenger gets off and waits (all the cars are idle with empty cars at this time). Another new military management device using fuzzy theory is the "elevator military management device" presented at the Battery and Information Management Society meeting held in October 3-5, 1988 in Nagana University, Japan. Decentralized pages 2, pp. 117-120). The following example is described here.

For example, when a car call is made in the upper floor and a certain car A is allocated, if the car is concentrated in the upper floor, the car can be assigned as a candidate car excluding this car A or a car having this property. do.

Also, a simulation example as shown in FIG. 19 is described. In this urban state, when the descending call on the 10th floor is registered, the car (unit 3) with the best evaluation value is selected as the allotment car among units 1 and 3 that are receiving car calls to the upper floor, except units 2 and 4, which are empty cars. It is stated that it is desirable to do so. According to this, the allocation in consideration of the occurrence of a call in the near future can be realized, and the occurrence of a long wait time can be prevented. In this simulation, however, the concentration of the car's upper floors is judged only as a condition of "calling the landing at the upper floor" or "assigning calls to the upper floor", and does not consider the change in the positional relationship between the cars with time. Because of this, one car may be called the upper layer, but since the positional relationship with other cars is not clearly determined, there is a possibility that the actual cars are not concentrated on the upper layer. In this case, the waiting time may be longer because the empty car is excluded from the allocation.

The above-mentioned problem is explained in more detail with reference to the example shown in FIGS. 20-22. In FIG. 2, the first unit A is traveling upward through the fourth floor in response to the car call 6c on the sixth floor, while the second unit B has no call on the sixth floor and is closed by an empty car. In this state, it is assumed that the falling call 9d has been registered on the 9th floor as shown in FIG. According to the fuzzy theory described above, when car B is assigned to the downcall of the 9th floor, it is determined that cars A and B concentrate on the upper floor, so that cars other than B (car A in this example) are set as candidate cars. The 9th floor falling call is assigned to the 9th floor. Therefore, if the rising call (1u) is registered in the lower floor, for example, immediately after the falling call (9d) of the 9th floor is allocated, the rising call (1u) of the first floor is assigned to the car B waiting on the 6th floor. Car B quickly moves to the first floor. Therefore, excessive long wait time does not occur. However, as described above, the effective case of excluding the vinca is only when the platform call is in a tendency to occur relatively frequently in the lower floor, and in the case of a fairly intermittent operation without a platform call in the lower floor for the time being, the following method is used. Is advantageous to minimize. That is, Vinca B responds to the 9th floor descent call, and for the landing call that occurs after a while, for example, the descent call 9d on the 9th floor is not shown in Fig. 22 unless it is assigned to the car A. As shown, Car A will wait 15 seconds after completing the operation on the 6th floor, so the car A, which will be free at that time, responds to the platform call.

As described above, in the conventional method, since the empty car is excluded from the allocation process without considering the probability of occurrence of the car (or predicted empty car) that will respond to the landing call in the near future and wait in the closed state, the effect of shortening the waiting time is insufficient. There is a problem. The onset of the problem is to solve the above problems, and aims to provide both an apparatus and a method for military management of an elevator that can shorten the waiting time associated with the platform call in the near future at this time.

According to an embodiment of the present invention for achieving the above object, a platform call registration means for registering a platform call generated by operating a platform button, selecting a car to be operated from a plurality of cars for the platform call, and then assigning the selected car. The car's driving direction, the car's starting direction, the car's starting, stopping and door control, and the car control means for responding the car to the corresponding car call and platform call, and the car after all car calls have been answered. On the basis of the car position predicted by the car position predicted means and the car position predicted means for predicting the position and direction of each car after a predetermined time has elapsed in response to the car call and the platform call. Detects cars that are expected to be in a state where there is no call to answer when a predetermined time elapses The elevator includes a predicted vehicle detecting means and an allocation limiting means for restricting a method in which the predicted vehicle detected by the predicted vehicle detecting means is assigned to the platform call by the assigning means when the landing call is registered in the landing call registration means. Provide military management devices.

According to another embodiment of the present invention, an elevator group management device includes a platform call registration means for registering a platform call generated when a platform button is operated, a car to be serviced for a platform call among a plurality of cars, and then the selected car to the platform call. Allocating means to be allocated, car driving direction such as car start / stop and car door opening / closing, car control means for responding the car to the corresponding car call and platform call, and the car after all car calls have been answered. Waiting means for placing the vehicle in a standby state, empty car detection means for detecting a closed car as a vacant car, and a car for predicting and calculating the position and direction after a predetermined time in response to a car call and a platform call After the predetermined time elapses based on the position predicting means and the car position and direction predicted by the car position predicting means Prediction vehicle detection means for detecting predicted vehicle, and when the vehicle detection means detects one or more vehicles or the predicted vehicle detection means cannot detect one vehicle at all, the platform call registration means is registered in the platform call registration means. In this case, at least one bin car detected by the bin car detecting means is limited to a method of allocating the landing call by the assigning means.

An elevator group management method according to another embodiment of the present invention includes a staff for inputting a status signal and a platform button signal transmitted from a car control apparatus, and a staff for registering a platform call based on the status signal and the platform button signal; A staff that detects closed cars without a call to be answered, a staff that calculates the estimated arrival time for each car to arrive at the platform, a staff that calculates the position and direction of each car after a predetermined time elapses; A staff that calculates the number of bin cars expected to be located in each zone after a predetermined time, the number of cars predicted, and the number of cars predicted to be in a state where there is no call to answer, and when the platform call is registered. The staff which temporarily allocates platform calls to each car based on the number of cars, the number of predicted cars and the number of predicted cars, and each car , Based on the result of evaluating the latency from the allocated landing call registration and staff, the staff rating consists of the staff assigned to the selected car to be allocated to the landing call. Next, an embodiment of this invention will be described with reference to the drawings. 1 to 10 are diagrams for explaining an embodiment of the present invention. The following description of this embodiment assumes that four cars are installed in a 12-story building.

1 is a block diagram showing the overall configuration of a group management apparatus for an elevator according to a first embodiment of the present invention, which is a car control apparatus 11 installed to control a group management apparatus 10 and Nos. 1 to 4 cars, respectively. ) To (14). 10A is a platform call registration means which registers and cancels platform calls (rise calls and descending calls) on each floor, and calculates the elapsed time, that is, the elapsed time, after the landing call is registered. 10B is an expected arrival time calculating means, and calculates an estimated value, i.e., an expected arrival time, before each car arrives at a platform (by direction) of each floor. 10C selects and assigns one car that is most suitable for serving the platform call as the allocation means, and assigns the operation based on the estimated waiting time of the platform call and the determination result by the allocation limiting means described later. 10D is a car position predicting means, and predicts the position and direction of the car after the predetermined time T hardening at the present time. 10E is a car number predicting means, and predicts the number of cars to be located in a predetermined platform area after the elapsed time T of the car predicted by the car position predicting means 10D.

10F is a waiting means, waiting for the car at the platform or the specific floor where the car is located when the car has answered all the calls. 10G is a vacant car detecting means, which detects a closed car as vacant car with no call to answer. 10H is a predicted empty car detecting means, and detects a car that is expected to become a empty car after a predetermined time T elapses among the cars that have been answered or are responding to the call. 10J is an allocation limiting means, which determines whether to limit or exclude the assignment of landing calls to the predicted vehicles. 11A is a well-known platform call canceling means, which is provided in the car control device 11 of No. 1 and outputs a platform call cancel signal for the platform call transmitted from each floor. The car control device 11 also controls the known car call registration means 11B for registering the platform call transmitted from each floor, and the known arrival forecast for controlling the lighting of the arrival line etc. (not shown) installed in each platform. Means 11C, a known driving direction control means 11D for determining and controlling the driving direction of the car, and a known driving direction for controlling the starting, running and stopping of the car in order to respond to the call or the assigned landing call. The control means 11E and the well-known door control means 11F which control opening and closing of an entrance door are included. Although not shown, the control devices 12 to 14 of Nos. 2 and 4 are also the same as those of the car control device 11.

2 is a block diagram of the military management device 10 and the military management device 10 includes a microprocessor (MPU) 101, a ROM 102, a RAM 103, an input circuit 104, and an output. It consists of a microcomputer (hereinafter referred to as a "microcom") that includes a circuit 105. The input circuit 104 is inputted with the boarding point button signal 19 transmitted from the boarding point buttons on each floor and the status signals of the 1st and 4th hogs from the car control devices 11 to 14, respectively. The output circuit 105 outputs a signal 20 from the boarding point built into each boarding point and a command signal to each of the control devices 11-14 for Nos. 1-4. Next, the operation of this embodiment will be described with reference to FIGS.

First, the outline of the military management operation of FIG. 3 will be described. The input program of the staff 31 includes the car position, direction, start, travel, stop, door open / closed state, car load, car call, and platform call cancellation from the platform button signal 19 and the control devices for the first to fourth cars. This is a well-known program for inputting a status signal. The platform registration program of the staff 32 is a well-known program used to determine whether or not to register or cancel the platform call and whether the platform button is turned on or off, and this program calculates the elapsed time of the platform call. The empty car detecting program of the staff 33 detects a car in a closed state as a empty car, as shown in FIG. In Fig. 4, the staff 46, i.e., the staffs 42 to 45, shows a procedure for detecting whether the first car is currently empty or not. Unit 1 does not have a car call or assigned landing call and is non-directional and the closed state proceeds to step 42 → step 44 → step 45 and sets blank car flag AV ₁ to “1” in step 45. do. Other than the above, the staff 43 resets the empty car flag AV ₁ to "0". Steps 47, 48, and 49 are sequentially judged as to whether the second to fourth cars are empty or not, and the empty car flag is set to the required level.

In the estimated arrival time calculation program of the staff 34, the estimated arrival time calculation program Aj (i) required for each car j (j = 1, 2, 3, 4) to arrive at each landing i. In this case i = 1, 2, 3,... 11 denotes each layer B2, B1, 1... Display the ascending platform on the 9th floor, while i = 12, 13,... 22 represents each layer 10, 9,... , 1, descent platform installed in B1. The estimated time of arrival is calculated by assuming that it takes 2 seconds to proceed to the first floor and 10 seconds to 1 stop, driving for one week on the Kaga train platform. The calculation of the expected arrival time is well known.

Position prediction step (35) the program in the first 4 hoka hoka predetermined time T has elapsed after the predicted car position P ₁ (T) ₄ ~P (T) and the predicted car direction D ₁ (T) ~D predict ₄ (T), respectively Calculate This is explained in detail with reference to FIG.

In the car position program 35 of FIG. 5, the staff 65, i.e., the staffs 51 to 64, has the predicted car position P ₁ (T) and the predicted car direction D ₁ (T) of the _first car after a predetermined time T elapses. Indicates the order of operation. When there is a platform call assigned to the first car, the process proceeds from the staff 51 to the staff 53. The staff 53 predicts the terminal floor in front of the most-allocated landing platform call as the final call floor of No. 1 car and considers the arrival direction of the car (falling direction at the top floor and upward direction at the bottom floor) at that floor. Set as call prediction platform h ₁ .

In addition, in the case of waiting for only the car call without the boarding point call to which the first car is assigned, the process proceeds in the order of the steps 51, 52, 54. In step 54, the farthest floor calling for the first car is predicted as the final call floor of the first car, and this final calling layer is set as the final call prediction platform h ₁ in consideration of the direction of arrival of the car at that time. In addition, even if the landing call is assigned to No. 1 car, even if there is no car call, the process proceeds in the order of Step 51 → Step 52 → Step 55. In the staff 55, the car position layer of No. 1 car is predicted as the final call layer, and the car call direction at this time is also taken into consideration and set to the final call prediction platform h ₁ .

In this way, when the final call prediction platform h ₁ is obtained, the process proceeds to the staff 56, where it is determined whether the first car is vacant. If No. 1 car is not Vinca (AV ₁ = “0”), go to step 57 to obtain an estimate of the time required for the No. 1 Car to become Vinca (hereinafter referred to as “Binka forecast time”) t ₁ . . From this, t ₁ is obtained by adding the predicted stop time Ts (= 10 seconds) of the stop time to the _first floor of the No. ₁ car to the h1 floor to the estimated time A ₁ of the arrival of the No. 1 car to the final call prediction platform h ₁ . When the car position floor is set to the final call prediction platform h ₁ by the staff 55, the remaining time is estimated by the stop state by the car state, for example, driving, deceleration, opening operation, and closing operation. Is set to the vinca prediction time t ₁ .

Next, in the staff 58, it is determined whether or not 1 car becomes a vacant car before a predetermined time T elapses. When the first car's empty car prediction time t ₁ is less than or equal to the predetermined time T, it means that the first car becomes an empty vehicle before the predetermined time T passes, and thus, the step 58 moves from the staff 58 to the staff 59, where the final call prediction platform h ₁ The displayed platform is set to the predicted car position F ₁ (T) at which the _first car reaches after a predetermined time T has elapsed. The prediction car direction D ₁ (T) is set to "0". The predicted car direction D ₁ (T) indicates car direction nothing, “1” indicates upward direction, and “2” indicates downward direction. The staff 60 sets the predicted empty car flag PAV ₁ of the first car to "1".

On the other hand, when the vacant car predicting means t ₁ of No. ₁ car is longer than the predetermined time T, it means that the vacant car has not yet been made even after the predetermined time T has elapsed. estimated arrival time a ₁ (i) of the i-1) estimated time of arrival a ₁ (i-1) and the landing (i) for this, {a ₁ (i-1) + Ts≤T <a ₁ (i) + The platform i indicated by Ts} is set to the predicted car position F ₁ (T) at which the first car reaches the predetermined time T. The prediction car direction D ₁ (T) is set in the same direction as the landing site i.

The step 62 resets the predicted blank car flag PAV ₁ of the first car to &quot; 0 &quot;. In step 56, when the first car is a vacant car (AV ₁ = "1"), the process proceeds to the step 63, and the vacant car prediction time t ₁ is set to 0 seconds. The next step 64 sets the platform indicated by the final call prediction platform h ₁ to the predicted car position F ₁ (T) at which No. 1 car reaches the predetermined time T. The prediction car direction D ₁ (T) is set to "0". Next, the staff 62 resets the predicted empty car flag PAV ₁ of the first car to "0".

In this way, the step 65 calculates the predicted car position F ₁ (T) and the predicted car direction D ₁ (T) for the _first car and detects the predicted empty car. Subsequently, the ₂ -PAV ₄ degree steps from the predicted car positions F ₂ (T) to F ₄ (T), the predicted car directions D ₂ (T) to D ₄ (T), and the predicted bin car flag for the ₂ car ₄ car, Steps 66 to 68 which are the same stop as 65) are set respectively.

Returning to FIG. 3 again, in the number of cars predicted by the staff 36, the number of cars located in a predetermined floor or a predetermined floor area after a predetermined time T elapses, for example, as shown in FIG. predicted car number with respect to each layer zone Z ₁ ~Z ₆ to the N ₁ (T) calculates the ~N ₆ (T), respectively. This car number prediction program will be described in detail with reference to FIG.

In the car number prediction program of FIG. 6, the step 71 sets the predicted car number N ₁ (T) to N ₆ (T) as “0” and initializes the car number j and the area number m to “1”. Set it. The staff 72 determines whether j hokka is located in the zone Zm after a predetermined time T has elapsed based on the predicted car position Pj (T) of the j hokka and the predicted car direction Dj (T). If it is predicted that j hokka is located in the zone Zm, the process proceeds to the staff 73 where the number of predicted cars Nm (T) of the zone Zm is increased by one. The staff 74 increases the car number j by one and proceeds to the staff 75 where it is determined whether or not the determination of the whole car is completed. If the determination is not finished, the process returns to the step 72 and the above-described processing is repeated. After all the cars at the staff 72 and the stem 73 have been treated for the zone m indicated by the zone number m, the staff number 76 is increased by one zone number m and the car number j is initialized to “1”. Set it.

The processing of the staffs 72 to 75 is repeated until the car number j> 4. Upon completion of the above-described processing for all the zones Z _{1 to} Z ₆ , the staff 77 determines whether the area number m> 6 or not, and when m> 6, the car number prediction process ends. The staffs 78 to 84 count the logarithmic NAV of the vinca and the logarithmic NPAV of the predicted vinca. The staff 78 initially sets the logarithmic NAV and the logarithmic NPAV to "0" and the number of lakes to "1", respectively. If j hokka is vinca, it progresses to the staff 79-80, and increases the number of vinca NAVs by one. If j hokka is the predicted empty car, the process proceeds from the staff 79 to the staff 81, and the staff 82 increases the number NPAV of the predicted empty cars by one staff. The staff 83 increments the car number j by one, determines whether the determination of all the cars has been completed by the staff 84, and returns to the staff 79 when the determination is not finished, and repeats the above-described processing.

In this way, the predicted number of cars N ₁ (T) to N ₆ (T), the number of cars NAV of the number of cars, and the number of cars of prediction number NPAV are counted, and the process of the number of cars predicted program 36 is completed. In the military management program of FIG. 3, the quota limiting program for the staff 37 is newly registered as the landing call C. The position of the landing call C is the layered position, and the predicted number of cars N ₁ (T) to N ₆ (T) at that time; On the basis of the number of vehicles NAV and the number of predicted numbers of NPV NPAV, it is determined whether or not the number of the No. 1 to No. 4 cars is assigned to the landing call C, and the allocation of the No. 1 to No. 4 cars to the new landing call C is made from the allocation system. and it sets the evaluation value P ₁ ~P _6. This allocation limit evaluation value P _{1 to} P ₄ means that as the value increases, the degree of allocation restriction increases, that is, when the value reaches infinity, it is excluded from the allocation object from the beginning. This determination procedure will be described in detail with reference to FIG.

In the allocation restriction program of FIG. 7, (1) the new landing call C belongs to the landing call group that occurs in the upper floor zone (Z ₃ or Z ₄ ), and (2) there is no vacancy (NAV <1), (3 ) The number of cars expected to be located in the upper layer zone (Z ₃ or Z ₄ ) for a predetermined time T is as many as N ₃ + N ₄ (T) ≤N _a , and (4) The number of predicted cars is 1≤NAPV≤N ₅ (5) If the landing call tends to occur frequently in the lower floor zone (Z ₁ or Z ₆ ), then processing proceeds in the order of staff 85 and staff 90 and predicted by staff 90; The quota limit evaluation PK for vinca K (k∈ {1, 2, 3, 4}) is set to “9999”, while car n (n∈ {1, 2, 3, 4} and n ≠ K are not predicted. } The allocation limit evaluation value Pn is set to "0".

In the above description, Na and Nb are constants. If at least one of the above conditions is not satisfied, all allocation limit evaluation values P _{1 to} P ₄ in step 91 are set to "0". In this way, the allocation limit evaluation values P _{1 to} P ₄ are set. In the third degree the group manager (10), the waiting time evaluation program of the step (38) Evaluation of the air to each landing call at the time when each provisionally assigned the new landing call C in 1 hoka 4 hoka time value W ₁ Calculate ˜W ₄ . It will be briefly described because it is known with respect to the calculation of the waiting time evaluation value W ₁ ~W _4. For example, in the case of temporarily allocating 1 car, if it is not registered in the platform call D (i) (i = 1, 2,…, 22) And the sum of their squared values, that is, the evaluation of the waiting time, N ₁ = U (1) ² + U (2) ² +. It calculates by + U (22) ² .

In the next step 39 allocation car selection program of FIG. 3, one allocation car is selected based on the allocation limit evaluation values P _{1 to} P ₄ and the waiting time evaluation values W _{1 to} N ₄ . In this embodiment, the total valuation value Ej at the time of temporarily allocating a new landing platform call C to j hoc is obtained by Ej = Nj + K · Pj (K: integer), and the car whose regular valuation value Ej is minimum is a regular allocation car. To choose. In the allocation car, an allocation command and a forecast command corresponding to the landing call C are set.

In the step 40 standby operation program of FIG. 3, if a vacant car ends in response to all platform calls, the vacant car is to be waited on the final call floor as it is, so that the car is not concentrated in one location, or at a specific floor. It is determined whether to wait, and when it is determined to wait on a specific floor, a waiting command for driving to the specific floor is set in the vacant car. In the step 41 output program of FIG. 3, the boarding point button signal 20 set as described above is sent to the boarding area, and an assignment signal, a forecast signal, and a standby command are sent to each car control device for No. 1 car and 4 car. In this manner, the group management program staff 31 to 41 are repeatedly executed.

Next, the operation of the military management program 10 of this embodiment will be described in more detail with reference to FIGS. 9 and 10. And for the sake of simplicity, explanation will be made by using FIG. 15 to FIG. 18. In the state of FIG. 15, it is said that the 9th floor falling call 9d is registered like FIG. The elapsed time of the rise call 5u of the third floor is assumed to be 10 seconds. In this case, if the downcall (9d) on the 9th floor is temporarily assigned to the car A, the waiting time for the downcall (9d) on the 9th floor and the upcall (5u) on the 5th floor is 24 seconds and 16 seconds, respectively. Value WA = 24 ² +16 ² = 832.

On the other hand, when the downcall (9d) on the 9th floor is temporarily assigned to the car B, the estimated waiting time of the downcall (9d) and the upcall (5u) on the 9th floor are 28 seconds and 16 seconds, respectively. Value W _B = 28 ² +16 ² = 1040. Therefore, in the typical snail allocation method, since W ₁ &lt; W _B , the 9th floor descent call 9d is assigned to the car A. After the predetermined time T (T = 20 seconds), the positions of cars A and B will be A ₁ and B ₁ in FIG.

Therefore, in each zone Z _{1 to} A ₆ , the predicted number is N ₃ (T) = 2, N ₁ (T) = N ₂ (T) = N ₄ (T) = N ₅ (T) = N ₆ (T) Since N = O, NAV = O and the predicted number of cars is NPAV = 1. In this example, the non-directional car is regarded as the upward direction, but may be arbitrarily determined in the driving direction of the car according to the car position. In this example if the predetermined value N = 2, _a = 1, as dae Nb refers to N3 (T) = if any car position to the second stand one station.

Therefore, the set to 7 degrees staff 90 quota to a program, a quota estimate PA = 99999 of the car A, car B the quota one evaluation value Z _B P _B = O for.

Therefore, the total valuation is E ₁ = D _B as E _A = N _A + P _A = 832 + 99999 + 100831, E _B = W _B + P _B = 1040 + 0 = 1040 Car B is assigned. In the conventional allocation method, since the car is assigned to the car A by the total evaluation value, the car is easily arranged in a line and an excessively long waiting call is generated.

However, in the above-described embodiment, the car arrangement can be prevented by considering the car arrangement after the predetermined time T has passed.

As is apparent from the above embodiment, in the above embodiment, the car sequentially predicts the car position and the car direction after a predetermined time elapses in response to the call, and the number of predicted cars is based on these and the car after the predetermined time elapses in each zone. Predictive operations are performed on the number of vehicles, and according to the number of predicted cars, the cars are not restricted to one location. Therefore, the car does not concentrate on one location, and the waiting time for the platform call can be shortened in the near future.

In the above embodiment, when the car position and the car direction are predicted after a predetermined time T, the car completes the response to the final call and first predicts the floor to which the empty car and the time required until then are based on these. Position and car direction were predicted. This is because the car is assumed to be waiting on the floor when the car becomes empty on a certain side.

If it is decided to always wait for a particular car on a particular side, it is assumed that the car always travels towards that floor, and the car position and the car direction are predicted. In addition, the car position and the car direction can be predicted by considering the new call that will occur until the predetermined time T elapses.

In addition, the above embodiment simplifies the calculation method of the final call prediction platform, but it is possible to make a more accurate and concrete prediction based on statistics calculated the probability of occurrence of the car call or the platform call.

In the above embodiment, the single building is divided into zones as shown in FIG. 8, but the division method is divided by various elements such as the number of floors, the number of cars, the time zone, or the use of each floor, for example, a main floor, a restaurant floor, a meeting room floor, and a transfer floor. It is easy to change.

Also, it is not necessary to decide the area considering the car direction arriving at each floor. Furthermore, in the above embodiment, when (C1) the new floor call (C) belongs to the platform group occurring in the upper floor area, (1) there is no empty car, and (2) the car bed that is expected to be located in the upper floor area after a predetermined time T has elapsed. (3) the number of predicted vehicles is between one and one Nb, and (4) if the platform calls occur frequently in the lower tier area, the allocation limit for allocating new platform calls C to the predicted vehicles is limited. Set each value (> 0).

However, the conditions for setting the allocation limit evaluation value are not limited to the above examples, and, for example, when the new landing call C belongs to the landing call group occurring in the upper floor area, the stamina of the condition (C ₂ ) is also possible. (C ₂ ) condition is (1) there is no empty car. ( ^* 2) The number of cars expected to be in the lower floor area after a certain period of time T passes. (3) The predicted binka number is between 1 and Nb. (4) Platform calls frequently occur in upper floor spaces.

Other conditions may be adopted when the new landing call C belongs to the landing call group that occurs in the mezzanine zone (C ₃ ). (C ₃ ) condition is (1) there is no empty car. (2) A large number of cars are expected to be located in the mezzanine zone after a certain period of time R. (3) The predicted binka number is between 1 and Nb. (4) Platform calls occur frequently in the upper or lower floor spaces.

In addition, if a particular location car is concentrated, a new landing call C may be registered (C ₄ ). (C ₄ ) condition is (1) no empty car. (2) Upper-floor space One of the lower-floor and middle-floor spaces, where the new platform call C does not belong, and there is no car that is expected to arrive after a predetermined time T. (3) The predicted binka number is between 1 and Nb. (4) Platform calls frequently occur in areas meeting (2) above.

If the condition (C ₄ ) is satisfied, the allocation of the prediction vehicle is preferably limited to the allocation of the prediction vehicle that is closest to the area satisfying the above (2).

Although the number of vacant cars is selected to 0 in each of the above conditions (C ₁ ) to (C ₄ ), the present invention can also be applied to the case of ≥ 1 blanks.

For example, when new landing call C belongs to the landing call group that occurs in the upper floor zone, (C ₅ ) conditions, that is, (1) one vacant car is located in the upper floor zone. (2) The number of cars expected to be located on the upper floor after a predetermined time T has elapsed. (3) The number of predicted binca is more than one, but less than a certain value Nb. Once this is established, the assignment of the new landing platform call C to a predetermined predicted car may be restricted.

This condition (C ₅ ) is particularly effective except in the case where a lot of platform calls continue to occur in the lower story. Specifically, the use of condition (C ₅ ) not only permits the rapid control of the vacant cars to be served in the new landing call C, but also the service of the soon to be registered vacant cars in the landing call to be registered in the lower tier area. I'm praying. As described above, each term (3) of the above conditions (C ₁ ) to (C ₅ ) states that the number of predicted blanks is one or more and less than a predetermined value Nb. This is because no allocation limit is required.

In the process of eliminating the condition of “less than Nb” and selecting the predicted vehicle to be allocated, the car which will be the fastest vehicle selected using one or two specific cars, for example, the empty car prediction time t _{1 to} t ₄ Alternatively, the same effect can be obtained by allocating the car closest to the selected lower floor zone using the predicted car positions F ₁ (T) to F ₄ (T) and the predicted car directions D ₁ (T) to D ₄ (T). In addition, various considerations can be taken on the basis of the number of predicted cars, which can be adopted as a condition for limiting the allocation of platform calls to the predicted cars. This condition can be easily realized as in the case of the condition (C ₁ ) shown in FIG. It is obvious. In addition, one or more conditions may occur at the same time when a plurality of conditions according to each situation are provided in addition to the above conditions (C ₁ ) to (C ₅ ). In such a case, there arises a problem of selecting a car to be assigned according to the condition.

Several methods have been proposed to solve this problem, and the well-known method is to use sebum theory. For example, it is described in the above-mentioned "Elevator Military Management Device" (1988 Institute of Electrical and Computer Information, 2nd Decentralized Book, 2nd page 117-120). In this method, the degree to which each condition constituting a predetermined condition is satisfied is represented by a membership function as a value between 0 and 1, and these values are used to select the minimum value for AND coupling and the maximum value for OR coupling. Based on the result of this selection, the confidence level of the condition itself is calculated, and the one with the highest confidence level is selected.

Alternatively, a weighting method is assigned to the limiting method of each condition according to the magnitude of the degree of confidence, and a limiting value is assigned to the car by the weighting result.

It is evident that the invention is also applicable to conditions in this manner. Thus, any condition can be selected as long as it is a condition to use the predicted number of cars to be concentrated in the predetermined floor or the predetermined floor area and the predicted number of cars to be filled in the near future.

The embodiment described above uses a method composed of the following staff.

In other words, as a means of limiting the allocation to the platform call, the total limit value is calculated by setting an allocation limit value of a specific car larger than the other cars, and adding the waiting time value by adding a weight to this value. An in-car is selected as an optimal car and assigned to a platform call.

In this way, the total evaluation and allotment of the allocation limit evaluation value in combination with other evaluation values means that the allocation evaluation value and the small car are allocated first. That is, cars with a large allocation limit evaluation value are more difficult to allocate than other cars. Further, the means for restricting the allocation to the platform call is not limited to the above embodiment, but may be a method of excluding a car that meets the allocation restriction condition from the car candidate group to be allocated in advance. In addition, in the above embodiment, the sum of squares of predicted waiting times of calls to which the waiting time evaluation values are to be raised or lowered is not limited thereto. For example, it is obvious that the sum of the predicted waiting times of a plurality of registered platform calls can be applied to the present invention, or similarly, a method of using the maximum predicted waiting time as the waiting time estimate can be applied to the present invention. Do. Of course, the evaluation item to be combined with the allocation limit evaluation value is not limited to the waiting time, and may be combined with an evaluation index which is used as a prediction error and a likelihood evaluation item. In the above embodiment, the car position and the car direction after one kind of predetermined time T have been predicted for each car, and the allocation limit evaluation value is calculated based on this, but other methods can be applied. For example, a plurality of different predetermined times T ₁ , T ₂ ,... , Tr (T ₁ <T ₂ <… <Tr) is installed to predict the position and direction of each car after each predetermined time, and located in each zone zm (m = 1, 2,…) after each predetermined time. The number of prediction cars to be performed is calculated, respectively Nm (T ₁ ) -Nm (Tr). The expected number of spaces NPAV (T ₁ ) -NPAV (Tr) is also calculated.

And each combination {N ₁ (T ₁ ), N ₂ (T ₁ ),. , NPAV (T ₁ )}, {N ₁ (T ₂ ), N ₂ (T ₂ ),. , NPAV (T ₂ )},. {N ₁ (Tr), N ₂ (Tr),... , The allocation limit evaluation values P (T ₁ ), P (T ₂ ),... , Add weight to P (Tr) and add it. In other words, P = K ₁ · P (T ₁ ) + K ₂ · P (T ₂ ). It is also easy to set the final allocation limit evaluation value P by calculating in accordance with the formula + KrP (Tr) (where K ₁ , K ₂ ... Kr is a weighting coefficient).

In this case, not only the car position of the specific time point T is evaluated, but the plurality of time points T ₁ , T ₂ . As a result, the Tr will be evaluated as a whole, which may further increase the waiting time for the platform call, which is a period that will elapse from now.

As shown in FIG. 11, the weighting coefficients K ₁ , K ₂ . The setting of Kr is for example at each time point T ₁ , T ₂ ,... Therefore, various setting methods are considered according to how the car is placed in the Tr, and the enemy may be selected according to the driving condition of the car and the characteristics of the building.

12 shows an allocation restriction method according to another embodiment of the present invention.

In Fig. 12, the same reference numerals as in the above embodiment denote the same parts.

The configuration of the other embodiment is the same as the one embodiment, except that the allocation limit program shown in FIG. 12 is used in place of the allocation limit program of the staff 37 of the military management program of FIG. In one embodiment, if there is no vacancy (NAV> 1), then the allocation limit for the predicted vacancy runs the step 37 allocation restriction program of FIG. 7, while in other embodiments there is at least one car (NAV≥1). The assignment of platform calls to predicted vehicles or vehicles is limited by the presence of predicted vehicles. Next, a method of restricting allocation according to another embodiment will be described with reference to FIG. The processing of the allocation limit program in FIG. 12 proceeds in the order of Step 92 → Step 93 → Step 94 → Step 95 → Step 96 under the following conditions.

(1) New landing call C belongs to the landing call group that occurs in the upper floor zone (Z ₃ or Z ₄ ).

(2) There is more than one vinca (NAV≥1)

(3) The number of cars expected to be located in the upper layer zone (Z ₃ or Z ₄ ) after the predetermined time T has elapsed is as large as N ₃ (T) + N ₄ (T) ≥Na.

(4) There is more than one predicted vinca (NPAV≥1).

In step 96, the allocation limit evaluation value PK for the predicted vehicle K (K∈ {1, 2, 3, 4}) is set to “99999”, while the car n (n∈ {1, 2) of the predicted vehicle is set. , 3, 4} and n ≠ K) are set to "0". Except for the case described above, the allocation limit evaluation values P ₁ -P ₄ for all cars in the staff 98 are set to "0".

In this way, the allocation limit evaluation values P ₁ -P ₄ are set.

Next, the operation of the military management program according to another embodiment will be described in more detail with reference to FIGS. 13 and 14. The example of FIGS. 20-22 is used again for simplicity of description. In the state of FIG. 20, it is said that the downcall 9d of the 9th floor is registered as FIG. At this time, when the ninth-floor downcall 9d is temporarily assigned to the car A, the predicted waiting time of the ninth-floor downcall 9d is 24 seconds, and the wait time evaluation value W _A = 24 ² = 576. On the other hand, when the falling call (9d) of the 9th floor is allocated to car B, the waiting time for the falling call (9d) of the 9th floor is 6 seconds, and the waiting time evaluation value W _B = 6 ² = 36.

After a predetermined time T (T = 20 seconds), cars A and B are positioned as indicated by cars A ₂ and B ₂ in FIG.

Therefore, the predicted number of cars becomes N ₃ (T) = 2, N ₁ (T) = N ₂ (T) = N ₄ (T) = N ₅ (T) = N ₆ (T) = 0, so the number of blank spaces NAV = 0, the predicted number of cars is NPAB = 1.

In the above example, the non-directional car is regarded as the above-mentioned direction car, but the direction of car travel may be arbitrarily determined according to the position of the car. In addition if, as in this example, _{Na = 2 N 3 (T)} = 2 , so tteuthage that position all cars one reverse versus the quota in a review of 12 ° step (96) the car A in a quota system program value P _A is "99999 ”, The allocation limit evaluation value P _B of car B is set to“ 0 ”. Therefore, the total valuation is E _A = W _A = W _A + P _A = 576 + 99999 = 100575 and E _B = W _B = P _B = 36 + O = 36, where E _A > E _B, so the 9th-floor downcall (9d ) Is eventually assigned to carB. In another embodiment, when the new landing call C belongs to the landing call group occurring in the upper floor zone, the condition (C ₆ ) is (1) a large number of cars expected to be located in the upper floor zone after a predetermined time T. There is more than one Vinca, and (3) there is more than one Forecast car.

When all of the above items (1) to (3) are satisfied, (4) an allocation limit evaluation value (> 0) for limiting the allocation to the new landing call C for the predetermined predicted vehicle is set.

In addition, when the new landing call C belongs to the landing call group that occurs in the upper floor zone, the condition (C ₇ ) is (1) the number of cars expected to be located in the upper floor zone after the passage of time T, and (2) one car As mentioned above, (3) There is not one predicted vehicle.

If all of the above items (1) to (3) are satisfied, (4) the allowable allocation limit evaluation value (> 0) for the new landing call C is set for the predetermined predicted vehicle.

However, the conditions for setting the allocation limit evaluation value are not limited to those described above. For example, the following conditions (C ₈ ) and (C ₉ ) are also acceptable.

(C ₈ ) when the new platform call C belongs to the platform call group that occurs in the lower floor area is (1) the number of cars expected to be located in the lower floor area after the passage of time T, and (2) more than one binka (3) There is more than one predicted vehicle.

If all of the above items (1) to (3) are satisfied, (4) an allocation limit evaluation value (> 0) is set for restricting the allocation to the new landing call C for the predetermined predicted vehicle.

In addition, when the new landing call C belongs to the landing call group that occurs in the lower zone, the condition is (C ₉ ) that (1) the number of cars expected to be located in the lower floor area after the passage of time T, and (2) Binka 1 (3) There are no predicted vehicles.

If the above conditions (1) to (3) are satisfied, (4) an allocation limit evaluation value (> 0) is set for limiting the allocation to the new landing call C for the predetermined vehicle.

In addition, the following conditions (C ₁₀ ) and (C ₁₁ ) may be adopted.

When the new platform call C belongs to the platform call group that occurs in the middle floor area, the condition (C ₁₀ ) is (1) the number of cars expected to be located in the middle floor area after the passage of time T, and (2) more than one binka. 3) There is more than one predicted vehicle.

When all of the above conditions (1) to (3) are satisfied, (4) an allocation limit evaluation value (> 0) for manufacturing the allocation for the new landing call C is set for the predetermined vacant car, respectively.

In addition, when the new platform call C belongs to the platform call group that occurs in the middle floor area, the condition (C ₁₁ ) is (1) a large number of cars expected to be located in the middle floor area after the passage of time T, and (2) one car (3) There is one prediction vehicle.

If the above conditions (1) to (3) are satisfied, (4) an allocation limit evaluation value (> 0) is set for limiting the allocation to the new landing call C for the predetermined vehicle.

Other If the number of cars concentrated in a specific location, the following conditions (C _12), and (C ₁₃₎ a may be possible to adopt the conditions when the new landing call register (C _12): (1) upper section, the lower layer zone, And any one of the mezzanine zones, which does not belong to the new platform call C, and has no Kaga, which is expected to arrive after a certain period of time T. (2) More than one Vinca, and (3) One predicted Vinca. Nothing is wrong.

If the above conditions (1) to (3) are established, (4) an evaluation value (> 0) is set for producing an allocation for the new landing call C according to a predetermined predicted vehicle.

In addition, when registered in the new landing call C, the condition (C ₁₃ ) is (1) any one of the upper floor, lower floor and middle floor areas, which does not belong to the new landing call C, and is expected to arrive after a predetermined time T. There are no areas, (2) one or more binca, and (3) no predicted binca.

If all of the above conditions (1) to (3) are satisfied, (4) an allocation limit evaluation value (> 0) is set for restricting the allocation to the new landing call C for the predetermined prediction car.

In addition, when there are a large number of predicted cars or vacant cars that meet the conditions (2) or (3) of each of the conditions (C ₆ ) to (C ₁₃ ), it is not necessary to maintain the predicted cars to be allocated. In this case, however, one or two specific cars, for example, the empty car prediction time t ₁ -t _4, are used to select the fastest car or the predicted car position F ₁ (T). It is preferred to limit allocation by selecting the car closest to the lower tier zone using -F ₄ (T) and predicted car directions D ₁ (T) -D ₄ (T).

In addition, the conditions of the allocation restriction for the vacant car or the predicted vacant car can be considered in various ways, and it goes without saying that the conditions C ₆ and C ₇ described in FIG. ₇ can be easily realized.

Claims (24)

Platform call registration means for registering the platform call generated when the platform button is operated; Allocating means for selecting and allocating a car to be serviced from a plurality of cars for the landing call; Car control means for controlling the driving of the car, such as driving direction, starting, stopping and opening / closing of the door, and responding the car to the corresponding car call and the platform call; Car position prediction means for predicting the car position and the car direction after a predetermined time in response to the car call and the platform call, and the car position and the car direction predicted by the car position prediction means; Predictive vehicle detecting means for detecting a predicted empty car which is not expected to be answered within a predetermined time; and when the landing call registration means registers the landing call, the predicting means detecting means detects by the assigning means. An elevator group management device, comprising: an assignment limiting means for restricting or excluding from the target of allocation of the predicted vehicle to a landing call. The elevator group management apparatus according to claim 1, wherein the car position predicting means calculates the car position and the car direction in consideration of a call expected to occur during the predetermined time elapsed period. The elevator group management apparatus according to claim 1, further comprising a bin car detecting means for detecting a car in a closed state, which has no call to answer, as a bin car. 4. The group management apparatus of an elevator according to claim 3, wherein the allocation limiting means restricts an allocation when the empty vehicle detecting means detects no empty cars. 5. The method of claim 4, wherein the limiting means is a platform call occurring in the upper floor area, the number of cars expected to be located in the upper floor after the predetermined time elapses, and the predicted vehicle detected by the predicted vehicle detecting means. An elevator group management device for restricting the allocation when the number is between 1 and a certain value. 6. The group management device of an elevator according to claim 5, wherein the limiting means restricts the allocation in a driving state in which platform calls are frequently generated in the lower floor area. 5. The method of claim 4, wherein the limiting means is that the landing call occurs in the lower floor area, the number of cars expected to be located in the lower floor area when the predetermined time elapses, and is detected by the predicted empty car detecting means. An elevator group management device for restricting the allocation when the number of predicted vehicles is between 1 and a certain value. 8. The group management device of an elevator according to claim 7, wherein said limiting means restricts said allocation in an operating state in which platform call is frequently generated in an upper floor area. 5. The method according to claim 4, wherein the allocation limiting means has a large number of cars expected to be located in the middle floor area when the landing call occurs in the middle floor, and the predetermined time has elapsed, and the prediction detected by the predicted vehicle detecting means An elevator group management device for restricting the allocation when the number of binka is between 1 and a certain value. 10. The group management device of an elevator according to claim 9, wherein the limiting means restricts the allocation in an operation state in which the landing call frequently occurs in an upper floor section and a lower floor section other than the middle floor section. 5. The method of claim 4, wherein the limiting means is a kaga that will be located when the landing call occurs in any of the zones of the upper floor, the lower floor, and the middle floor, and the predetermined time has elapsed into an area outside the area where the landing call occurred. An elevator group management apparatus for restricting the allocation when there is an unexpected area and the number of predicted cars detected by the predicted cars detecting means is between 1 and a certain value. 12. The group management apparatus of an elevator according to claim 11, wherein the limiting means restricts the allocation in a driving state in which a landing call frequently occurs in the area where there is no prediction car. 4. The method according to claim 3, wherein the limiting means has a large number of cars that are expected to be located in the upper floor area when the landing call occurs in the upper floor area, the empty car detecting means detects the empty car, and the predetermined time elapses, and the An elevator group management apparatus for restricting the allocation when the number of predicted cars detected by the predicted cars detecting means is between 1 and a predetermined value. 2. The method according to claim 1, wherein the car position predicting means predicts the car position and the car direction for each predetermined time elapse of a different kind, and the predicted vehicle detecting means predicts a predicted bin for each predetermined time elapsed. And an operation for detecting a car, wherein the allocation limiting means performs an operation for limiting the allocation in consideration of the predicted empty car detected by the predicted empty car detecting means for each of a predetermined time elapse of the different kind. Platform call registration means for registering a platform call generated when operating a platform button, assigning means for selecting and allocating a car to be serviced for the platform call among a plurality of cars, and driving direction, start, stop and door opening and closing of the car. Car control means for controlling a car driving and for responding the car to the corresponding car call and the landing call; A vinca detecting means for detecting a closed car as a vacant car, a car position predicting means for predicting car position and car direction after each predetermined time in response to a car call and a platform call, Predicted vehicle detection means for detecting a car, which is expected to have no call to be answered within the predetermined time, based on the car position and the car direction predicted by the car position prediction means, as the predicted vehicle, and the vehicle detecting means comprises: If the landing call registration means registers the landing call when one or more empty cars are detected and the predicted vehicle detecting means cannot detect the predicted empty car at all, the one or more empty cars detected by the empty car detecting means by the allocating means. Eli consists of a means of allocating restrictions which restricts or excludes the allocation of cars to platform calls. The group management apparatus of the data. 16. The group management apparatus of claim 15, wherein the car position predicting means calculates the car position and the car direction in consideration of a call that is expected to occur during the predetermined time elapsed time. 16. The apparatus according to claim 15, wherein the allocation limiting means assigns one or more empty cars detected by the predicted vehicle detecting means to the landing call by the assigning means when the predicted vehicle detecting means detects one or more empty vehicles. Military management device of the elevator to limit the way. 18. The group management device of an elevator according to claim 17, wherein the limiting means restricts the allocation when the platform call occurs in the upper floor section and the number of cars expected to be located in the upper floor section after the predetermined time elapses. 18. The group management apparatus of claim 17, wherein the limiting means restricts the allocation when the landing call occurs in the lower floor area and the number of cars expected to be located in the lower floor area after the predetermined time has elapsed. 18. The group management device for an elevator according to claim 17, wherein the limiting means restricts the allocation when the platform call occurs in the middle floor area and the number of cars expected to be located in the middle floor after the predetermined time elapses. 18. The method of claim 17, wherein the means for limiting allocation occurs in any one of the upper, lower and middle layers, and there is an area without a car to be located when the predetermined time has elapsed into an area other than the area where the landing call occurred. An elevator group management apparatus for restricting the allocation when the number of predicted vehicles detected by the means is between 1 and a predetermined value. A step of inputting a platform button signal and a status signal transmitted from the car controller; A step of registering a landing call on the basis of the landing button signal and a status signal; A staff that detects empty cars that have no calls to answer and are closed; Staff estimating the estimated arrival time for each car to arrive at each floor; A staff for calculating a predicted position and direction of each car after a predetermined time elapses; A staff that calculates the number of bin cars and the number of predicted cars expected to be located in each zone after a predetermined time, and the number of predicted cars that are not expected to be answered; A staff for temporarily allocating a landing call to each car based on the number of bin cars, the predicted cars number and the predicted cars number when the landing call is registered; A staff for evaluating a dash time from the registered landing call assigned to each car; And a step of selecting and assigning a car to be allocated to the landing call based on the result of the evaluation step. 23. The method of claim 22, wherein the temporary allocation staff does not have the vacancy when the predetermined time has elapsed, the number of predicted vacant cars is between 1 and a certain value, the number of cars expected to be located in a zone, and the landing call is The group management method of the elevator which restricts the allocation at the same time, in the operating state where the platform call frequently occurs in an area outside the said area at the same time. 23. The method of claim 22, wherein when the predetermined time has elapsed, there are one or more vacant cars, a large number of cars expected to be located in a zone, and one or more predicted vacant cars when the platform call occurs. The group management method of an elevator which restricts the allocation to a vacant car and does not restrict the allocation to the vacant car if there is no said predicted vacant car.

Images